Parallel Bayesian Optimization of Multiple Noisy Objectives with Expected Hypervolume Improvement

Abstract

Optimizing multiple competing black-box objectives is a challenging problem in many fields, including science, engineering, and machine learning. Multi-objective Bayesian optimization (MOBO) is a sample-efficient approach for identifying the optimal trade-offs between the objectives. However, many existing methods perform poorly when the observations are corrupted by noise. We propose a novel acquisition function, NEHVI, that overcomes this important practical limitation by applying a Bayesian treatment to the popular expected hypervolume improvement (EHVI) criterion and integrating over this uncertainty in the Pareto frontier. We argue that, even in the noiseless setting, generating multiple candidates in parallel is an incarnation of EHVI with uncertainty in the Pareto frontier and therefore can be addressed using the same underlying technique. Through this lens, we derive a natural parallel variant, $q$NEHVI, that reduces computational complexity of parallel EHVI from exponential to polynomial with respect to the batch size. $q$NEHVI is one-step Bayes-optimal for hypervolume maximization in both noisy and noiseless environments, and we show that it can be optimized effectively with gradient-based methods via sample average approximation. Empirically, we demonstrate not only that $q$NEHVI is substantially more robust to observation noise than existing MOBO approaches, but also that it achieves state-of-the-art optimization performance and competitive wall-times in large-batch environments.

arXiv commentTo appear in Advances in Neural Information Processing Systems 34, 2021. 40 pages. Code is available at https://github.com/pytorch/botorch

References in corpus (17)

• Predictive Entropy Search for Efficient Global Optimization of Black-box Functions
• A Tutorial on Bayesian Optimization
• An Easy-to-use Real-world Multi-objective Optimization Problem Suite
• Bayesian Optimization with Unknown Constraints
• BoTorch: A Framework for Efficient Monte-Carlo Bayesian Optimization
• On the representation of the search region in multi-objective optimization
• Predictive Entropy Search for Multi-objective Bayesian Optimization with Constraints
• Maximizing acquisition functions for Bayesian optimization
• Parallel Bayesian Global Optimization of Expensive Functions
• Real-world Video Adaptation with Reinforcement Learning
• A Flexible Framework for Multi-Objective Bayesian Optimization using Random Scalarizations
• Random Hypervolume Scalarizations for Provable Multi-Objective Black Box Optimization
• Fast Matrix Square Roots with Applications to Gaussian Processes and Bayesian Optimization
• Parallel Predictive Entropy Search for Multi-objective Bayesian Optimization with Constraints
• Thompson Sampling for Contextual Bandit Problems with Auxiliary Safety Constraints
• Modeling an Augmented Lagrangian for Blackbox Constrained Optimization
• Distilled Thompson Sampling: Practical and Efficient Thompson Sampling via Imitation Learning

Cited by in corpus (8)

• Multi-Objective Bayesian Optimization over High-Dimensional Search Spaces
• Pareto Front-Diverse Batch Multi-Objective Bayesian Optimization
• Ranking over Regression for Bayesian Optimization and Molecule Selection
• Latency-Aware Neural Architecture Search with Multi-Objective Bayesian Optimization
• Perceptual-Distortion Balanced Image Super-Resolution is a Multi-Objective Optimization Problem
• Carbon-Efficient Neural Architecture Search
• Experimenting, Fast and Slow: Bayesian Optimization of Long-term Outcomes with Online Experiments
• Advancing the Pareto front for thin-film materials using a self-driving laboratory